Data sources
Flows
Function
Flows (formerly Reload Flows) define when and in which order data sources and dataflows are reloaded. They are useful whenever a data source or dataflow has to be loaded after another one finishes, or when reloading needs to follow a schedule.
Flows can be used with any pull data source that supports a reload interval. Event-based sources such as MQTT or OPC UA cannot be reloaded by a Reload step, because they decide on their own when to update. They can, however, be used as a trigger: with the After data reload trigger, an incoming MQTT message or OPC UA value change starts the flow.
A flow is built from these elements
| Trigger | Decides when the flow runs. Four triggers are available today: Periodic (sec) for a fixed interval in seconds, Schedule for a weekday/time schedule, After data reload, which fires the flow as soon as a specific data source has finished reloading, and After function execute, which fires the flow as soon as a specific function has run. |
| Step | A unit of work executed by the flow. The Reload step reloads any data source or dataflow. The Run function step executes a Lua function you created beforehand in the package explorer — for example to compute values, prepare data, or trigger follow-up actions. Steps run sequentially in the order shown. |
| Failure function | An optional Lua function that runs when a step fails. Useful for logging, alerting, or running compensating actions. |
Note
The same data source or dataflow may be used in multiple flows. Be careful not to create endless loops — for example by using an After data reload trigger that reloads the very data source it listens to.
Creating a flow
To create a new flow, click the [Project] menu in the [Home] tab of the Peakboard Designer (1) and then click [Flows] (2).
As long as no flows exist, the Manage flows dialog shows an empty state with two entry points: the [+] icon in the left column (1) and the central [Add flow] button (2). Either one opens the editor for a new flow.
Configuring the flow
In the editor, first give the flow a name (1). Then drag the building blocks from the [Select triggers and steps] panel on the right into the highlighted drop zones in the middle area:
- A trigger from the right-hand trigger picker (4) onto the yellow drop band in the Triggers section (2). In the example, Periodic (sec) is set to fire every 60 seconds. A flow can have one or more triggers.
- A step from the right-hand step picker (5) onto the drop band in the Steps section (3). The Reload step then shows a dropdown where you pick the data source or dataflow to reload — test in the example.
The order of the steps defines the execution order — you can rearrange them at any time by dragging. To remove a trigger or step, use the trash-can icon at the right end of the row.
Note
A flow without any trigger only runs when it is called manually, for example from a Lua script (flows("MyFlow"):run()).
Click [OK] to save the flow, or [Cancel] to discard the changes.
Preparing data sources for flows
Before a flow can reload a data source, only the [Enabled] checkbox (1) in the Reload section of the configuration dialog needs to be turned on. That is all that is required: as soon as the data source is used in a Reload step, the flow’s trigger decides when it is reloaded.
The mode selected under [Reload state] (2) and the [Interval (in s)] field (3) only control the data source’s own independent reloading. When it is reloaded by a flow, the flow trigger overrides that interval — a periodic schedule set here has no effect. In practice it is therefore enough to set [Reload state] to Manual and leave all timing to the flow.
Running functions in a flow
Besides data sources, a flow can also execute your own functions. The function has to be created first in the package explorer under [Scripts] › [Functions] (1) — only then is it available in the flow.
Then, in the flow editor, drag the Run function step (2) from the right-hand step picker into the Steps section and pick the function from the dropdown (1). In addition, the After function execute trigger (3) is available to start a flow as soon as a specific function has run.
Editing or deleting existing flows
In the Manage flows dialog, pick the flow you want to change from the list on the left. The three-dot menu next to the flow name lets you rename or delete the flow. Triggers and steps can be removed by hovering over the entry and clicking the trash-can icon that appears.
Tip
The optional Failure function lets you attach a Lua function that runs whenever a step in the flow fails. This is a clean way to write failed reloads to a log data source or fire an alerting push notification.
